Simulation and measurements of the effects of different urban environments on GPS location errors using digital elevation models and building databases

A simulation model will be described and employed to study the propagation channel of the Global Positioning System. Initially, a digital elevation model, building databases, and a vegetation model are processed to generate azimuth-elevation maps of path states (clear, shadowed and blocked) for a large number of observers. At each simulation step, satellite positions are updated using an orbit predictor and azimuths and elevations of paths from satellites to observers calculated. Signal strengths and range errors are assigned to paths with the aid of a random number generator for each path state. This information is processed to determine the cumulative distribution functions for position errors. It will be shown that, by proper selection of parameters, model predictions are able to display a good agreement with measurement results. The simulation model will then be applied to a large number of observers deployed along two routes in densely urbanized areas in the City of Rio de Janeiro (22.8°S, 43.3°W) with buildings displaying different height distributions to show how the position errors change with the average building height. In combination with the comparison between measurement and prediction results, this indicates that the simulation model may be a useful tool for studying and planning satellite-based location and navigation applications with good accuracy and sensitivity.